Missense mutations in small muscle protein X-linked (SMPX) cause distal myopathy with protein inclusions

Figures

• 

  Fig. 1. a Pedigrees of the families included in the study. DNA samples were collected from the individuals marked with an asterisk. Corresponding gentoypes are displayed for these individuals where Xm represents a mutated allele and X represents a wild-type allele. F7 II.3 is indicated as suffering from poliomyelitis and thus was not included in the study. b Haplotype analysis, showing the Haplotype A (Italian/Maltese haplotype) observed in Families F1–F2 and Haplotype B (French haplotype) observed in F6–F7 along with the corresponding age of the haplotypes. c A summary of the identified missense mutations in SMPX and their correlation with the observed phenotypes in F1–F9

• 

  Fig. 2. Magnetic resonance imaging (MRI) T1 sequences of patients with SMPX distal myopathy. a Patient F1 II.1 at 57 yrs showing normal thighs but fatty degeneration in lower legs: anterior compartment and medial gastrocnemius muscles. b F2 II.2 at 61 yrs showing minor degenerative change in thigh muscles semimembranosus, biceps femoris and left vastus intermedius; fatty degenerative changes in the anterior compartment muscles more on the left of proximal lower leg and fatty replacement of anterior compartment and part of the soleus muscles in the distal lower legs. c F3 II.1 at 58 yrs (left lower limb) with normal thigh and severe fatty degeneration in lower leg anterior compartment, medial gastrocnemius and medial part of distal soleus. d F6 II.1 at 52 yrs showing normal thigh and fatty degenerative changes in the anterior compartment (more on the right) and medial gastrocnemius muscles of proximal lower leg, and anterior compartment with soleus muscles in the distal lower legs. e F6 II.1 at 60 yrs with early degenerative changes in biceps femoris and semimembranosus in the thigh and more fatty degeneration of anterolateral compartments and of medial gastrocnemius and soleus in the…

• 

  Fig. 3. Histochemical and DAB immunohistochemical stainings. a Hematoxylin & eosin (HE) staining shows fiber size variation and internal nuclei, multiple cytoplasmic inclusions are present showing sarcoplasmic and subsarcolemmal localization (arrows). b, c Rimmed vacuoles (arrows) are observed in HE (b) and immunostaining for the autophagosome marker LC3b (c). d In Gomori’s trichrome, sarcoplasmic inclusions show red labeling. e, f Enzyme histochemical COX-SDH e and mitochondrial NADH diaforase f stainings show small irregular sarcoplasmic areas of reduced activity. g Myofibrillar protein accumulation as shown by myotilin staining is present in several fibers in DAB immunostaining, whereas desmin h does not accumulate in a similar manner. i–l In p62/SQSTM1, ubiquitin and SMI-31 stainings, multiple positive protein inclusions and punctate labeling are observed in nearly all fibers, whereas TDP-43 staining remains negative. Panels f–l are serial sections. Scale bars = 100 µm

• 

  Fig. 4. Immunofluorescence microscopy. a–f Immunofluorescent double staining; merged images are shown, SMPX is green and CRYAB is magenta. In control skeletal muscle (ctrl) free of neuromuscular disease, SMPX shows diffuse cytoplasmic and focal subsarcolemmal staining pattern. Muscle biopsy of F1 II.1 b shows moderate SMPX accumulation in a single fiber (white arrowhead), whereas in patient F2 II.2 c there is no SMPX accumulation. In probands from families F7, F8 and F9 (d–f), multiple SMPX-positive sarcoplasmic inclusions are present in several fibers. In F8 III.2 (e) and F9 II.1 (f), separate myofibrillar CRYAB accumulation is observed (white arrowheads), which does not co-localize with SMPX labeling. g–i Double immunostaining of F6 II.1. Single channel and merged images are shown; SMPX is green, vinculin is magenta. SMPX-positive protein inclusions (g) are also positive for vinculin (h), merged image (i). j–l; serial sections Myofibrillar accumulation (arrow) is positive for myotilin (j), and also for the CASA proteins BAG3 (k) and HSPB8 (l). m–n; serial sections) Several (atrophic) fibers show sarcoplasmic up-regulation of both BAG3 (m) and (n) HSPB8. Scale bars = 100 µm

• 

  Fig. 5. SMPX aggregation in patient muscle. a Congo red staining in fluorescence microscopy using Texas red filter showing abundant positive sarcoplasmic inclusions in muscle biopsy from patient F9 II.1 (scale bar = 50 µm). b Confocal sections of a muscle biopsy from patient F8 III.1. Immunostaining of SMPX or myotilin (green) combined with Amytracker 680 (AT680, magenta) shows Amytracker fluorescence in SMPX inclusions. A serial section of the same fiber shows no signal when Amytracker is omitted (BAG3/ctrl), showing that the protein inclusions are not autofluorescent. c–e EM findings of patient F9 II.1. c One large subsarcolemmal protein inclusion with accumulation of filamentous amyloid-like material (black arrowhead). d Myeloid bodies (white arrowhead) surrounding sarcoplasmic inclusions (black arrowhead). e Two classic cytoplasmic bodies (CB) with radiating filaments, together with six individual sarcoplasmic inclusions (black arrowheads) and nucleus (N)

• 

  Fig. 6. Aggregation propensity of mutant SMPX a In Pasta2.0 prediction, an N-terminal region of wild-type SMPX showed aggregation propensity, which was further enhanced by the p.P7T and p.A13V mutations, as indicated by the decrease in free energy. The gray dashed line shows the aggregation threshold of 5 PASTA units. b Representative fractionation experiment demonstrating reduced solubility of mutant SMPX-V5. Total protein staining of the V5 blot, and tubulin and histone 3 (H3) stainings of a separate blot are shown as loading and fractionation controls. UT, untransfected; N100 and N50, normalization samples. (c) Quantification of SMPX-V5 in supernatant and pellet fractions. The graph shows mean ± SD from three triplicate experiments (n = 9), normalized to total protein loading and to the mean of wild-type SMPX within each experiment. Asterisks indicate significant differences compared to WT (two-tailed Mann–Whitney U test with Bonferroni correction; * p < 0.05; ** p < 0.01)

• 

  Fig. 7. Stress granules. a HeLa cells showed cytoplasmic SMPX-V5 foci that were positive for TIAL1. Scale bar = 20 µm. For other SG components in non-treated and MG132-treated cells, see Supplementary Fig. 2 (online resource). b, c Quantification of TIAL1 SGs in cells positive for SMPX-V5 (gray bars) and V5-negative cells in the same wells (white bars) after MG132 (b) or arsenite (c) stress. NT, non-treated; MG, 20 µM MG132 2 h; ARS, 500 µM sodium arsenite 45 min; R1/R3, recovery of 1 h / 3 h after stress. The graphs show mean ± SD of group means from 4 experiments. Asterisks indicates a significant difference to WT according to Dunnett’s multiple comparisons test (p = 0.0038; mean diff. –0.1737; 95% CI –0.2424 to –0.1050). For details, see Supplementary Tables 2–3 (online resource). d Delayed recovery of SMPX-enriched SGs. SGs from the V5-positive cells in (b) were classified as V5-enriched (dark bars) or non-enriched (light bars) and expressed as SG number per cell. For details, see Supplementary Fig. 2 and Supplementary Table 4 (online resource)